Access to capital: Milwaukee's continuing small business lending gaps

A C C E S S T O C A P I T A L : M I L W A U K E E ' S C O N T I N U I N G S M A L L

BUSINESS LENDING GAPS

Gregory D. Squires and Sally O'Connor

Small businesses constitute the backbone of the U.S. economy. The roughly 24 million small businesses nationwide employ 52 percent of the private workforce, contribute 51 percent of private sector output, create the majority of new jobs, and produce 55 percent of innovations. These businesses, and the U.S. economy generally, depend on access to credit, particularly from commercial banks, in order to survive and thrive (Of- rice of Advocacy, U.S. Small Business Administration 1999).

Yet small businesses, and particularly small minority-owned businesses in urban communities, often experience difficulty in obtaining small business loans. Recent evidence indicates that minority-owned firms receive fewer and smaller loans than white-owned firms with identical traits (Cavalluzzo, Cavalluzzo, and Wolken 1999; Blanchflower, Levine, and Zimmerman 1998; Ando 1988; Bates 1989, 1997). Research conducted by the National Bureau of Economic Research found in a 1993 survey that black business owners were three times as likely as whites to be turned down for small business loans. Among firms that were comparable in terms of credit rating, age, size, geographic location, and experience of owners, black-owned firms were twice as likely to be rejected and among those firms that were approved, black ownership paid an average of one percent more in interest (Blanchflower, Levine, and Zimmerman 1998). In addition to these national studies, similar findings

This research was supported by the Helen Bader Foundation. We want to thank Donna Schenstrom, Head Cartographer of the Cartographic Services Laboratory at the University of Wisconsin-Milwaukee, for preparing the maps. We also wish to express our appreciation for data processing assistance provided by Frank Stetzer, Senior Consultant with Information and Media Technologies at the University of Wisconsin-Milwaukee.

10 The Review of Black Political Economy/Fall 2001

have been reported from local case studies of small business lending (Conta and Associates 1990; Immergluck and Mullen 1997; Immergluck and Wiles 2000).

Under the federal Community Reinvestment Act (CRA) depository institutions are required to affirmatively ascertain and be responsive to the credit needs of their entire service areas, including low--and moderate-income communities. New regulations promulgated in 1995 require large commercial banks and thrift institutions to report annually small business and small farm lending by geographic location to their regulators (12 C.F.R. w Marsico 1996). Lending activity for the 1996 calendar year was reported to federal financial regulatory agencies and in October 1997 small business and small farm lending data were released to the public for the first time by the Federal Financial Institutions Ex- amination Council (FFIEC). The most recent small business lending data currently available are for the 1999 calendar year.

This study provides a detailed review of small business lending in Milwaukee, Wisconsin and illustrates how this new data set can be utilized to assess small business lending in virtually any local market. Mil- waukee is a fairly typical industrial community that has been hit hard by decades of disinvestment but which also has been the location of many successful community reinvestment initiatives in recent years (Squires and O'Connor 2001). Previous research found that among the nation's fifty largest metropolitan areas Milwaukee had the smallest share of small business loans going to low--and moderate-income areas (Norman 1998). Lending to small businesses, that is firms with assets below $1 million, was also found to be below nationwide levels. Small business lending has also been concentrated in white communities with black and His- panic communities receiving relatively small shares of such loans and loan dollars. But lenders vary dramatically in Milwaukee in terms of the distribution of their small business loans by neighborhood income level (Squires and O'Connor 1999). This study examines changes in small business lending patterns in Milwaukee between 1996, when these data first became available, and 1999, the most recent data that are available.

DATA AND M E T H O D O L O G Y

Under the new CRA rules issued in 1995 banks with assets totaling more than $250 million or affiliated with a holding company totaling more than $1 billion in assets are required to report small business and small farm loan information to their regulatory agency beginning with

Squires and O'Connor 11

the 1996 calendar year. That information includes the number, dollar volume, and census tract of business loans for less than $1 million and farm loans up to $500,000. Lenders are also required to report lending activity to businesses and farms with annual revenues below $1 million, an approximation of lending to small businesses. In other words, not all small business loans go to small businesses.

Data made available by the FFIEC report aggregate lending (i.e. loans made by all lenders combined) by census tract by county. But individual lender disclosure reports do not provide data at the tract level. For individual lenders data are reported by aggregations of census tracts according to median family income levels of those tracts. The categories used in this analysis are the following: 1) low-income (median family income in the tract is less than 50 percent of the metropolitan area median family income); 2) moderate income (50-79 percent); 3) middle income (80- 119 percent); and 4) upper income (120 percent or more).

This study examines small business loans in the four-county Milwau- kee metropolitan statistical area (MSA). The four counties are Milwau- kee, Ozaukee, Washington, and Waukesha. Small business loans are de- fined as those whose original amounts are $1 million or less and which were secured by nonfarm or nonresidential real estate. Generally this includes loans that meet the definition of "loans to small business" that are reported in Call Reports and Thrift Financial Reports (Federal Finan- cial Institutions Examination Council 1996: 13).

Nationwide the number of reporting institutions declined from 2,078 in 1996 to 1,911 in 1999. But the number of reported loans increased from 2.4 million to 3.2 million and the number of reported loan dollars grew from $147 billion to $173 billion (Federal Financial Institutions Examination Council 1997 and 2000). Since smaller institutions are not covered by these reporting requirements, not all small business lending is included. But among depository institutions, these reports account for approximately two-thirds of all small business lending. These loans in- clude originations and purchases with originations accounting for more than 98 percent of the total in 1996 and 96 percent in 1999. A slight majority of these loans went to businesses with annual revenues below $1 million in 1996 with approximately 60 percent of these loans going to such small businesses in 1999 (Federal Financial Institutions Examina- tion Council 1997:5,6 and 2000: 9; Greenspan 1998: 3). In Milwaukee there were 15,181 small business loans totaling just under $1.5 billion with originations accounting for virtually all (99.8 percent) of them in 1996. The number of loans increased to 17,356 and the number of loan


dollars grew to just over $1.6 billion in 1999 with originations accounting for 99.6 percent in 1999. The share of loans going to small businesses declined slightly from 52.2 percent to 50.0 perecent.

There are several limitations to these data. First, the location reported for the borrower may not be the same location that is supported with the borrowed funds. A business may have several locations and some or all of the borrowed funds may be invested in neighborhoods other than the one that is reported to the federal regulator as the main address of the organization. Most businesses, however, have only one location so the extent of misclassification is minimal. Second, no information is provided on credit demand. That is, unlike mortgage loans reported under the Home Mortgage Disclosure Act (HMDA), there is no information on the number or types of businesses that applied for a loan. Consequently, there is no information on the disposition (e.g. approval or denial, rea- sons for denial) of applications for small business loans. Third, again unlike the HMDA data, there is no information on the race, gender, or income of those receiving business loans. Fourth, only loan originations and purchases are reported. Outstanding loans made in previous years are not included so total lending activity in a given area by a particular institution may not be fully accounted for in these reports. Finally, business lending activity is reported only by census tract. Once again unlike HMDA data, individual small business loan data are not reported. Given the limited information about the characteristics of borrowers and the nature of demand for small business loans as well as the limited information on the disposition of the loan and the nature of the supply for such credit, it is important to cautiously interpret any reported differences in the distribution of small business loans (Bostic and Canner 1998).

The following section examines aggregate small business loans and loan dollars by census tract income level for financial institutions doing business in the Milwaukee metropolitan area and compares those patterns to all reporting institutions nationwide in 1996 and 1999. The ratio of loans and loan dollars per person and per business are also presented by tract income level. Loans and loan dollars to businesses with annual revenues of less than $1 million by tract income level are then presented. Business counts were generated by Dun and Bradstreet and provided by the Board of Governors of the Federal Reserve System. Aggregate lending activity on the part of financial institutions serving the Milwaukee area is then examined by racial composition of neighborhoods. Finally, comparative data are provided for each lender that made more than 200 business loans in Milwaukee in both years. This sample includes 14


lenders which accounted for approximately 70 percent of all reported small business loans and loan dollars in the metropolitan area in both years.

FINDINGS

Four basic findings emerge from this analysis. First, lending activity in Milwaukee is concentrated in middle--and upper-income areas, and is more concentrated in such communities than is the case nationwide. These disparities have moderated, but only slightly, in recent years. Sec- ond, lending to small businesses (i.e. the proportion of all small business loans and loan dollars that went to firms with assets below $1 million) in Milwaukee is below nationwide levels, particularly in low-income areas, and these gaps have expanded in recent years. Third, small business lending is consistently concentrated in predominantly white communities with black and Hispanic communities receiving relatively small shares of such loans and loan dollars. More significantly, gaps between black and white communities have grown considerably. And fourth, Milwaukee area lenders vary substantially in their distribution of small business loans by neighborhood income level.

Small Business Lending by Neighborhood Income Level

Total small business lending grew much more slowly in Milwaukee than nationwide between 1996 and 1999. And in Milwaukee small business lending has consistently been concentrated in middle and upper income neighborhoods (see Figure 1) with lending in low--and moderate-income areas below national averages in 1996 and 1999. However, this gap has closed modestly during these years.

The number of small business loans in Milwaukee grew by 14.3 percent compared to 34.0 percent for the nation as a whole. The growth in small business loan dollars was 7.3 percent in Milwaukee and 17.4 percent nationwide (see Table 1). The share of loans going to low--and moderate-income census tracts was approximately 13 percent in Milwau- kee and 19 percent nationwide in these years. (These census tracts account for just over 25 percent of the population of the Milwaukee metropolitan area and just under 20 percent of all small businesses.) Compa- rable figures for loan dollars were just under 12 percent for Milwaukee and just under 22 percent for the United States (see Table 2). These numbers reflected a very slight increase in lending activity to these neigh-

14 The Review of Black Political Economy/FaU 2001

Squires and O'Connor IS

.3

o

o

5 o I.-

(J

~ o ~ c0~-

,,.I

,._-~

8 ~

~

t-~ t'N

d

i,j,

d ~

o o

~ 0 ~ 0

0

D,- t~l

0

QPO

0

~o

8

C'4

~4

c~

=o

~ g

t o

~ q

!

~ E = ~ 0

~ <


~4

.C

C J i

o) 0

~ q

~ ~ ,

i ~ e- o

i _= |~ o ~ - , -= ~-

O

r ~

E ~. o

~

o ' d

~o~

O O

�9 ~ o ~ . )

O

~ O


borhoods in Milwaukee and a slight decrease for the nation as a whole, but a substantial gap persists.

Ratios of loans per person and per business were approximately the same in Milwaukee and the U.S. generally, though loan dollars per person and per business were much higher in Milwaukee. Unfavorable gaps persist in low--and moderate-income areas though, once again, they ap- pear to be closing.

In Milwaukee, loans per 100 businesses grew from 32 to 37 but the increase was much greater nationwide going from 30 to 40. In low--and moderate-income areas loans per 100 businesses grew from 40 to 53 (a growth rate of 32.5 percent) in Milwaukee and from 50 to 57 (a growth rate of 14.0 percent) nationwide. Even among middle--and upper--income areas loans per 100 businesses were lower in Milwaukee than nationwide in 1999, but this was not the case in 1996. So gaps persist, though at a slightly smaller rate, in low--and moderate-income areas, and new gaps emerged in the wealthier neighborhoods.

Loan dollars per 100 businesses grew from $3,070 to $3,335 in Mil- waukee and from $1,816 to $2,131 for the U.S. So loan dollars per business were 69 percent higher in Milwaukee than the nation generally in 1996 and 56 percent greater in 1999. But this is not the case in l o w - - and moderate-income areas. For example, in Milwaukee's low-income census tracts the ratio of loan dollars per 100 businesses increased from $1,950 to $2,255 while nationwide the increase was from $1,800 to $1,903. So loan dollars per business were just 8 percent higher in Mil- waukee in low--and moderate-income tracts in 1996 and 18.5 percent higher in 1999 than in comparable neighborhoods nationwide. This did represent a 15.6 percent increase in Milwaukee compared to an increase of 5.7 percent nationwide in such communities (see Table 3 and Figure 2).

A slightly different picture emerges in examining loan dollars per 1,000 persons. This ratio was approximately $1,000 in Milwaukee in both years and between $582 and $684 nationwide. But in low--and moderate-income tracts this ratio was considerably smaller in Milwaukee and, in fact, was well below the national figures. In low-income areas this figure changed from $449 in 1996 to $484 in 1999 compared to a nationwide increase from $661 to $699. (By contrast, loan dollars per 1,000 persons exceeded $1,100 in middle-income areas and $1,400 in upper income areas in Milwaukee during these years compared to roughly $550 and $800 nationwide. See Table 4, Figure 2, and Maps 1 and 2). In other words, when comparing Milwaukee to the nation generally, Mil-


~.~ .~

d

~ 0

~ O O O

~ v ~ ~ , , ~

< <

r a


FIGURE 2 Small Business Lending

Milwaukee and the U.S., 1996 and 1999


F I G U R E 2 (continued) Small Business Lending

Milwaukee and the U.S., 1996 and 1999


g. g,

e~

3

. ~ i ~ o, ~ ~ ~ ' ~ -,~ ~.~ _~~ ~

E ~

o - d , - ~ . N

..~ N

8 < --- ~.)


MAP 1 Small Business Lending by Neighborhood Income Level

Milwaukee MSA 1999


MAP 2 Small Business Lending Change by Neighborhood Income Level

Milwaukee MSA 1996-1999


waukee fares better in loan dollars per businesses even in low--and moderate-income areas. But Milwaukee fares worse when the comparison is based on loans per persons. The fact that Milwaukee fares more favorably compared to the nation as a whole when businesses rather than persons is the point of comparison (and this is particularly the case in low--and moderate-income areas) suggests a lower level of entrepre- neurship in these communities in Milwaukee. That is there may be fewer small businesses, or fewer that qualify for credit, relative to the size of the population in Milwaukee than elsewhere.

Overall, small business lending in Milwaukee continues to be concentrated in middle--and upper-- income census tracts, and to a greater extent than is the case nationwide. But to some extent the gaps are moderating.

Loans to Small Businesses

The share of loans and loan dollars going to small businesses (i.e. firms with assets of $1 million or less) has been consistently smaller in Milwaukee than nationwide, and the gap increased between 1996 and 1999. These disparities were greatest in low--and moderate-income areas.

Total loans to small businesses grew by 9.7 percent in Milwaukee and 45.3 percent nationwide while loan dollars to small businesses grew by just 1.4 percent locally and 34.2 percent for the nation generally (see Table 5). The percentage of these loans that went to small businesses actually declined in Milwaukee by 2.2 percent while increasing nationwide by 4.7 percent. In Milwaukee the share of loan dollars that went to small firms dropped by 2.0 percent compared to a nationwide increase of 6.1 percent (see Table 6).

While these disparities persisted at all income levels, they were greatest in low--and moderate-income communities. The share of loans going to small firms dropped by 3.9 percent and 9.9 percent in low--and moderate-income areas, stayed virtually the same in moderate income census tracts, and dropped by 2.4 percent in upper income areas in Milwaukee. Nationwide, these figures increased in all areas with the increase being the greatest in low--and moderate-income areas. Nationwide the increases in low--and moderate-income areas were 8.9 percent and 5.7 percent compared to 3.4 percent and 5.4 percent in middle--and upper-income areas. The share of loan dollars going to small finns also declined in Milwaukee at each income level while for the nation generally the share of loan dollars going to small businesses increased at each income level.


c ~

.

3

3 0

m.

e4

0 ~

~ o~

I . o 0 ~o t - -

c

d ~e - & ~ ~ |

"0

8~

~~ c~

"0 0

x~

8<

a~

0


.3

~ N

~ -i

O O

O ~ c . / 3

e~

O 6./3


A relatively smaller number of small businesses or businesses that apply and qualify for credit in Milwaukee compared to the nation generally might account, in part, for these gaps in any given year. Why the share of loans and loan dollars to small businesses should decline over time relative to the nation, however, is more problematic.

Lending Activity By Neighborhood Racial Composition in Milwaukee

In the Milwaukee metropolitan area lending activity has consistently been concentrated in predominantly white communities. These gaps generally grew during these years, in some cases considerably.

The number of loans to predominantly white neighborhoods (where blacks constituted less than 10 percent of the population) increased by 8.9 percent but declined in black neighborhoods (where blacks accounted for more than 70 percent of the population) by 10.2 percent. Loan dollars in white areas increased by 4.6 percent and declined in black areas by 47.0 percent. By far the biggest increase was in areas that were between 10 percent and 70 percent black. Loans increased by 62.0 percent and loan dollars grew by 27.7 percent in these neighborhoods. Consequently, as a percentage of all loans, the share in predominantly white and black areas declined slightly. A different picture emerges for Hispanics. In both predominantly white (less than 5 percent Hispanic) and Hispanic areas (greater than 25 percent Hispanic) the number of loans increased about the same (13.9 percent and 12.1 percent) but the number of loan dollars increased faster in Hispanic areas (l 2.5 percent) than in white neighborhoods (5.9 percent--see Tables 7 and 8 and Figure 3).

More revealing are loans and loan dollars per 1,000 persons. The ratio of loans per 1,000 persons varied from 13 in predominantly white areas to 2 in predominantly black areas in 1996. This gap increased moderately to 14 to 2 in 1999. On this measure, lending activity was more than six times greater in white than in black neighborhoods in 1996 and seven times greater in 1999. Similarly, in 1996 white neighborhoods had al- most three times as much lending activity as Hispanic neighborhoods and approximately two and one-half times as much in 1999.

In examining loan dollars a different pictures emerge. The Hispanic gap remained at approximately 3 to 1 as the ratio of loan dollars per thousand persons in white areas compared to Hispanic areas went from $1,056/$330 to $1,126/$372. But for blacks the gap grew substantially. The white/black ratio of loan dollars per thousand persons went from 5.5 in 1996 ($1,192/$215) to 10.2 in 1999 ($1,247/$122). This reflected a

28 The Review of Black Political Economy/Fal l 2001

~ a

�9 , j

i Q

A ~

n ~

~ �9

~. o . ~

o4

o.

t~

0 0

o

to

o

to

to

o

IN

to

0

I,,-

cO t o cO ~ t o ~

0 (,9 ~ 0

~ M

to

to *-" r--"

r ,-"

._o ._o ._o r r e=

t'~

,r-

f--.

< . - :

,ZD O .-~ "G

o < ~<

- ~

S q u i r e s a n d O ' C o n n o r 29

0 ~ ~. ~. o. o 0

I g . "-

o~ ~0 cO

o 0 0

Q 0 0

o 9 o

~ o

o~ o T-

o ~, o

c4 ~d o o

._~ u ._o

0

("4

r~

< _ :

o < ~<

o


FIGURE 3 Small Business Lending by Neighborhood Racial Composition

Milwaukee MSA, 1996 and 1999


FIGURE 3 (continued) Small Business Lending by Neighborhood Racial Composition

Milwaukee MSA, 1996 and 1999


decline of 43.3 percent in the loan dollars per thousand persons in predominantly black areas (from $215 to $122) compared to an increase of 4.6 percent in white ($1,192 to $1,247) and 27.7 percent ($668 to $853) in integrated areas (see Table 9, Figure 3, and Maps 3,4,5 and 6).

Lending activity per business by racial composition could not be cal- culated because the data on the number of small businesses were not available at the individual census tract level. This information was made available only by aggregate census tracts based on income level. From the available data, it is impossible to determine the number of businesses in a group of census tracts characterized by their racial composition.

Racial gaps not only persist but, in comparing black and white communities, they have grown larger. On some measures they have moderated slightly for Hispanics. But on several key measures the white/black gap increased substantially.

Variations Among Milwaukee Area Lenders

Differences between Milwaukee area lenders and lenders nationwide pale in comparison to disparities among financial institutions the Mil- waukee metropolitan area. Just as some mortgage lenders have far sur- passed their colleagues in levels of service provided to lower-income communities (Squires and O'Connor 1998), business lending in these communities also varied dramatically among lenders.

As indicated above, just under six percent of loans and loan dollars went to low-income census tracts in Milwaukee in 1996 and 1999. L o w - - and moderate-income areas combined received approximately 14 percent of loans and 12 percent of loan dollars in these years. Among the 14 lenders who made at least 200 small business loans in 1996 and 1999, the percentage of loans going to low-income areas ranged from zero to 12.9 percent. Three lenders were below the 3 percent level in both years. Combining low--and moderate-income areas, the range was from zero to 27.5 percent (see Table 10). For loan dollars the range was from zero to 10.3 percent in low-income areas. Three lenders were at or below 1 percent. Combining tow-and moderate-income areas the range was from zero to 23.3 percent (see Table 11).

Rates of change varied as well. Five lenders had a higher percentage of their loans in low--and moderate-income areas in 1999 than was the case in 1996 with eight exhibiting declines. The largest increase in l o w - - and moderate-income areas was 12.2 percentage points (from 15.3 percent to 27.5 percent for Firstar) while the biggest drop was 17.9 percent-


o ~ ~=~

e~

~ e ~ , =~

m

O

e~ b E ' -

t -

o

o

p~

~ b

~D

o~

~ o

~ r

o

~D

~D

c~ 0o

r162

r

~o

~4

o~

e- ._o r

~ c ~ .

S~

o~

<

L~ o L ~


MAP 3 Small Business Lending By Black Population Concentration

Milwaukee MSA 1999


MAP 4 Small Business Lending By Hispanic Population Concentration

Milwaukee MSA 1999


MAP 5 Small Business Lending Change By Black Population Concentration



MAP 6 Small Business Lending Change By Hispanic Population Concentration



age points (from 26.7 percent to 8.8 percent for M & I Northern). For loan dollars six had an increase to low-income tracts and seven had increases to low--and moderate-income areas. The largest increase to low--and moderate-income areas was 8.8 percentage points (from 7.0 percent to 15.8 percent for M&I Marshall & Ilsley) while the steepest decline was 14.3 percentage points (from 23.3 percent to 9.0 percent for M & I Northern). One lender, M & I First National Bank, made zero loans to low--or moderate-income areas in both years.

Loans to small businesses varied as well, with the majority making between 50 percent and 80 percent of their loans to such businesses in both years. Most lenders made between 30 percent and 70 percent of their loan dollars to small firms in these years. Just five provided a larger share of both their loans and loan dollars to small firms in 1999 than in 1996 (see Table 12).

While aggregate data for all lenders reporting loan activity in the Mil- waukee area combined are available at the individual census tract level, as indicated above for each individual lender data are available only by aggregations of census tract according to median family income levels of those tracts. Therefore, it is not possible to determine lending activity for an individual institution by racial composition of neighborhood. But lending by income level and to small firms varies substantially among lenders. Given the association between income and race, it is likely that lending patterns also vary substantially by the racial composition of neighborhoods as well. In Milwaukee, access to small business loans appears to be at least in part a function of the institution to which a borrower applies. Characteristics of those institutions, along with the credit worthiness of borrowers, likely affect the distribution of small business loans.

The Woodstock Institute recently found that banks with branch offices in low--or moderate-income areas make a greater proportion of their loans in those areas (Immergluck and Mullen 1997). This appears to be the case in Milwaukee as well. In 1996 those lenders with branches located within economically distressed census tracts, referred to as the "Target Area" by the Comptroller of the City of Milwaukee (City of Milwaukee 1996: 21), made 6.6 percent of their loans in low-income tracts compared to 5.1 percent for lenders without a Target Area branch. Figures for loan dollars were 6.8 percent and 5.5 percent. But lenders without Target Area branches provided a slightly higher proportion of their loans and loan dollars to small firms (Squires and O'Connor 1999). More research on a range of lender characteristics would yield additional information on the factors influencing the allocation of small business

Squires a n d O ' C o n n o r 39

~a =

Z "N

i,,.- _ _ 0 i ,,,..

+! .,1

ID

P �9 g. m

~ T T � 9

~ ~ 0 ~ ~ 0

+ + + + . + + + + m + + ' + , + + ' + ' , + + _ , + ~ - , , ,

I/I

i ~. - z ~ .~ ~+~. .+" ,++ + + + - ..+ ,~,+ +- + +..++ -~ ~<+ -..- ++ p+ p+ + + =+=+:~+++ ,,,{I:

N

t.O

~ . o .

0

O 0

L ~

++-D

.=_+ o

,.+ ~ .+.. ~ . ~ o

0 ~ 2 ~ o

~ ~~-

.= .~ ,~ ~ ~

~ o ~


" ~ <

~ . - ~ = N

=

"0 ~" " �9

~ m m

&

O~

~ ~ 0 ~ 0 0 ~

~ ~ 0 ~ ~ 0 ~

� 9 ' ~ ' ' ~

14.

fM oO

t ~ v--

p.- 09

O O �9 o

[1. ~ .

~5

o9

O

2 ~

o ~

o

~ < r 0J

, " O [.~ o9 " ~ L~

~ o


S ~

.= N

"N

r~

o

Q~

L.

E

~ . , , -

0 I - - 'm

~ N ~ N N

d ~ N g M M N ~ d ~ N d d

0 ~ ~ 0 ~ ~ d ~ o ~ ~

0 ~ 0 0 ~ ~ 0 ~

= c ~ o~ u~: =~82~a~z ~ z ~'~

�9

c~

o

t-- 0%

0 <2

�9 ~ =~

o>

�9

(-q 0

%)


loans. Size of lender, whether it is independently owned or part of a regional or national holding company, working relationships with community organizations, and racial composition of the work force are just some of the factors that might affect the distribution of small business loans.

RESEARCH AND POLICY IMPLICATIONS

In the Milwaukee metropolitan area small business lending continues to be concentrated in upper-income and predominantly white communities. These patterns may reflect differences in demand, credit-worthiness of borrowers, unfamiliarity on the part of potential borrowers and lenders about prevailing opportunities, unlawful discrimination, Milwaukee's low rate of minority business ownership and representation in corporate man- agement compared to other metropolitan areas (Norman 1998) and a range of other factors. But these clearly are not random fluctuations.

These findings also reveal substantial, ongoing differences between Milwaukee area lenders and financial institutions nationwide in the distribution of loans and loan dollars by neighborhood income levels. Com- pared to their counterparts nationwide, Milwaukee area lenders provide a lower concentration of their lending activity in lower-income neighborhoods and among small businesses though some gaps have moderated. The findings also indicate widespread disparities among lenders in the distribution of small business loans throughout Milwaukee area neighborhoods. These patterns may reflect legitimate differences in marketing strategies among lenders, illegal discrimination against low-income areas and minority communities, business opportunities overlooked by some institutions, or a combination of these and other factors. The broad disparities suggest that something other than the quantity and quality of the demand for credit accounted for current lending patterns since all of these reporting institutions are serving the same metropolitan area.

Clearly, far more research is essential to fully understand the underly- ing causes and policy implications of these findings. But recent disclosure of small business lending provides additional insight into the lending behavior of financial institutions covered by the CRA. Three minor changes in current small business disclosure requirements would enhance the value of this information.

First, covered lenders should be required to report the number of applications for small business loans they receive along with the disposition of those applications. This would provide at least one measure of demand


for such credit and further insight into the response of lenders to that demand. This provision is included in the Community Reinvestment Mod- ernization Act introduced into the U.S. House of Representatives by Tom Barrett (D-Milwaukee) and Luis Gutierrez (D-Chicago) in March 2001.

Second, the Federal Reserve Board should act on a proposal it is currently considering that would allow lenders to solicit information on the race of small business loan applicants. Currently, requesting such information violates the Equal Credit Opportunity Act. The U.S. Depart- ments of Justice and the Treasury, the Comptroller of the Currency, and the Office of Thrift Supervision support the proposed regulatory changes which would facilitate collection of this information (Reno 1998). This change would also be required by the Community Reinvestment Mod- ernization Act.

Anecdotal and econometric evidence suggests that minority-owned businesses have more difficulty accessing small business loans than majority- owned firms. While disclosure of this information alone would not confirm or deny the existence of unlawful discrimination, it would enhance current understanding of racial disparities in access to small business loans. Many businesses, of course, are owned by more than one person. In those cases where there is multiple ownership, if more than 50 percent of the business is owned by members of a particular race, that would be the one which is reported. A multi-race option could also be utilized where no single group is controlling.

Such additional reporting could increase the chances that the identity of a particular business would be revealed. The FFIEC could establish a threshold for a minimum number of loans (e.g. at least three loans in a tract) that must be reported before the individual loan data would be revealed in order to preserve confidentiality where it might otherwise be breached.

Third, the FFIEC should release tract level data for individual lenders and make available tables that display lending activity by racial composition of tracts. Users could locate the racial composition of each tract and then aggregate them in order to examine the distribution of loans for all reporting institutions by neighborhood racial make-up, as has been done here. However, data currently available do not permit such analysis at the individual bank level. FFIEC should provide this information in a similar manner as mortgage lending data are provided in HMDA reports. Se- lected aggregate HMDA reports display lending activity (e.g. applications, originations, etc.) in tracts that are less than 10 percent minority, those between 10 and 19 percent, 20 to 49 percent, 50 to 79 percent, and


80 to 100 percent�9 These reports are available for individual lenders and for all lenders combined by metropolitan area. Similar tables reporting business loan activity should be prepared and disseminated by FFIEC. Again, where confidentiality might be breached, the information could be suppressed.

The data examined in this study and called for in these recommenda- tions, alone, would not be sufficient to confirm or deny the existence or prevalence of compliance or non-compliance with the CRA or other fair lending rules. Comparative analysis of individual loan files or paired testing by "mystery shoppers" posing as small business credit applicants would be required for that purpose. Testing has been used quite effec- tively to detect and remedy discrimination in a variety of areas including mortgage lending, property insurance, employment, and in a number of other everyday commercial transactions and public accommodations like service in restaurants, hailing a taxi, and access to hotel rooms (Fix and Turner 1999). Current testing methodologies could be modified to yield additional information on the level of discrimination in small business lending markets. But even in the absence of such investigative tech- niques, the patterns that are revealed from analyses of existing small business lending data can provide guidance to regulators, lenders, and others concerned with problems and potential opportunities in accessing such credit. These data can assist regulators in targeting and conducting their examinations, reveal potential trouble spots for lenders, and identify missed opportunities for financial institutions and their community reinvestment partners. The minimally expanded reporting and disclosure rec- ommended here can enhance the value of that guidance.

HMDA and CRA have changed the way many mortgage lenders do business and increased the supply of funds for community reinvestment (Evanoff and Segal 1996; Schwartz 1999; Shlay 1999; Squires and O'Connor 2000;)�9 Hopefully, the new CRA regulations calling for disclosure of small business loans can extend those effects by nurturing an increase in the availability of small business loans in previously underserved communities.

R E F E R E N C E S

Ando, Faith. 1988. "Capital Issues and Minority-Owned Business," Review of Black Political Economy 16 (4): 77-109.

Bates. Timothy. 1989. "Small Business Viability in the Urban Ghetto," Journal of Regional Science 29 (4): 625-643.

�9 1997. "'Unequal Access: Financial Institution Lending to Black--and White- Owned Small Business Start-ups," Journal of Urban Affairs 19 (4): 487-495.


Blanchflower, David G., Phillip B. Levine, and David J. Zimmerman. 1998, "Dis- crimination in the Small Business Credit Market," National Bureau of Economic Research Working Paper No. 6840.

Bostic, Raphael W. and Glenn B. Canner. 1998. "New Information on Lending to Small Businesses and Smalt Farms: The 1996 CRA Data," Federal Reserve Bulletin 84 (1): 1-21.

Cavalluzzo, Ken S., Linda C. Cavalluzzo, and John D. Wolken. 1999. "'Competition, Small Business Financing, and Discrimination: Evidence from a new Survey," in Jackson L. Blanton, Alicia Williams, and Sherrie L. W. Rhine (ed) Business Access to Capital and Credit Washington, D.C,: Federal Reserve System,

City of Milwaukee. 1996. Annual Review of Lending Practices of Financial Insti- tutions, Milwaukee: Office of the Comptroller, City of Milwaukee~

Conta and Associates, Inc. 1990. "A Study to Identify Discriminatory Practices in the Milwaukee Construction Marketplace," Milwaukee: Conta and Asso- ciates, Inc.

Evanoff, Douglas D. and Lewis M. Segal. 1996. "CRA and Fair Lending Regulations: Resulting Trends in Mortgage Lending," Economic Perspectives XX, (6): 19-46.

Federal Financial Institutions Examination Council�9 1996. "'Community Reinvest- ment Act Interagency Questions and Answers Regarding Community Reinvest- ment," Richmond: Federal Reserve Bank of Richmond, October 21.

�9 1997. "Findings from Analysis of Nationwide Summary Statistics for 1996 Community Reinvestment Act Data Fact Sheet," Washington, D,C.: Federal Fi- nancial Institutions Examination Council. September 30.

�9 2000. "Availability of Data on Small Business, Small Farm, and Commu- nity Development Lending," Washington, D,C.: Federal Financial Institutions Examination Council. August 29.

Fix, Michael and Margery Austin Turner (ed) (19991. A National Report Card on Discrimination in America: The Role of Testing Washington, D.C.: The Urban Institute.

Greenspan, Alan. 1998. "Remarks by Chairman Alan Greenspan at a Community Forum on Community Rein'~estmem and Access to Credm California's Chal- lenge, Los Angeles, California," January 12, http:llwww.bog.frb.fed.uslboarddocsl speeches/19980112.htm.

lmmergluck, Dan and Erin Mullen. 1997. "New Small Business Data Show Loans Going To Higher-Income Neighborhoods in Chicago Area," Chicago: The Woodstock Institute�9

Immergluck, Daniel and Marti Wiles. 2000. "Where Banks do Business: Small Busi- ness Lending Patterns in the Chicago Area, 1996-1998." Chicago: The Woodstock Institute.

Marsico, Richard D. 1996. "The New Community Reinvestment Act Regulations: An Attempt to Implement Performance-Based Standards," Clearinghouse Review March:. 1021-1030.

Norman, Jack. 1998. "City lowest in loans to firms in poor areas," Milwaukee Journal Sentinel May 25.

Office of Advocacy, U.S. Small Business Administration. 1999. "Small Business Lending in the United States, 1998 Edition," Washington, D.C,: U.S. Small Busi- ness Administration.

Reno, Janet�9 1998. Speech delivered before the Seventh Annual Conference of the National Community Reinvestment Coalition, March 20,


Schwartz, Alex. 1998. "Bank Lending to Minority and Low-Income Households and Neighborhoods: Do Community Reinvestment Agreements Make a Difference?" Journal of Urban Affairs 20 (3): 269-302.

Shlay, Anne B. 1999_ "Influencing the Agents of Urban Structure: Evaluating the Effects of Community Reinvestment Organizing on Bank Residential Lending Practices," Urban Affairs Review 35 (2): 247-278.

Squires, Gregory D. and Sally O'Connor�9 1998. Milwaukee's Best (and Worst?) Banks, Milwaukee: Fair Lending Coalition.

�9 1999. "Access to Capital: Milwaukee's Small Business Lending Gaps," in Jackson L. Blanton, Alicia Williams, and Sherrie L. W. Rhine (ed) Business Access to Capital and Credit Washington, D.C.: Federal Reserve System.

�9 2001. Color and Money: Politics and Prospects for Communi~ Reinvest- ment in Urban America Albany: State University of New York Press.

Documents

Access to capital: Milwaukee's continuing small business lending gaps